Machine Learning Model to Predict Urban Sprawl Using Official Land-use Data Cover Image

Machine Learning Model to Predict Urban Sprawl Using Official Land-use Data
Machine Learning Model to Predict Urban Sprawl Using Official Land-use Data

Author(s): Mohamed Noby, Mohamed E. Elattar, Omar Hamdy
Subject(s): Physical Geopgraphy, Regional Geography, Environmental Geography, Applied Geography, Policy, planning, forecast and speculation, Environmental interactions
Published by: INCD URBAN-INCERC
Keywords: urban growth; machine learning; similarity weight; Aswan;

Summary/Abstract: The rate of global urbanization is constantly increasing. As a result of the massive population growth, there is an increasing demand for further urban development, especially in developing regions such as Aswan city. This paper aims to examine the usage official land-use data in predicting future urban growth until 2046, moreover, to define urban driving forces in case study area. This was done using Similarity weighted model, a machine learning based model to simulate future urban growth. The results show that official land-use data produce a slightly better results’ accuracy than remote sensing sources within small to medium scales. The results although reveal that for study region, urban area is expected to expand to cover an area of almost 4460 Feddan by year 2046. The outcome of this research assesses decision makers to accurately predict future urban sprawl areas using available official land-use data.

  • Issue Year: 14/2023
  • Issue No: 3
  • Page Range: 249-258
  • Page Count: 10
  • Language: English
Toggle Accessibility Mode